Because of the health benefits associated with DHA and EPA, several international agencies and organizations have put together recommendations for EPA and DHA supplementation and for fish consumption . This initiative enabled not only health promotion but also the reduction of the risk of several chronic inflammatory diseases [3, 23]. Although much has been learn about n-3 FAs, many questions remain, including the dose-response effect on clinical outcomes as well as the differential effects on health of EPA and DHA.
The objective of the present study was to investigate the effect of different concentrations of EPA, DHA, and EPA + DHA on inflammatory markers, in stimulated THP-1 macrophages. It is well accepted that EPA and DHA are associated with a lower risk of inflammatory diseases . Their health effects are usually studied as a sum or combination. Thus, little is known about the potential health effect of EPA compared to DHA. Nevertheless, few studies have shown EPA- or DHA-specific effects on inflammation. Gorjao and collaborators reported different effects of EPA and DHA on endothelial cells, insulin-secreting cells, and leukocyte function . In a study on RAW 264.7 macrophages, Honda et al. demonstrated that EPA and DHA attenuated cell inflammatory activities and more importantly that their respective effect changed in potency depending on the investigated cytokines (IL6 and TNFA) . Finally, an investigation of the effect of EPA and DHA on THP-1 macrophages showed that EPA and DHA had a differential effect on cytokine transcription . It is important to further understand the specific effect of n-3 FAs on inflammatory mechanisms and to know whether they have complementary, shared, or divergent effects. A previous study by our research group  investigated the effect of 50 and 10 μM EPA, DHA, and EPA + DHA on non-stimulated macrophages. This study firstly showed that in non-inflammatory conditions, EPA and DHA had a varying effect on the expression of inflammatory genes and that the anti-inflammatory effect was dose-dependent .
In the present study, our ultimate goal was to investigate the anti-inflammatory effects of n-3 FAs in inflamed cells. In this perspective, the current study investigated the effect of EPA, DHA, and EPA + DHA on stimulated THP-1 macrophages. n-3 FA effects were tested on seven genes involved in inflammation. An incubation of these macrophages into 100 ng/ml LPS for 18 h allowed mimicking an inflammatory state. Stimulation with LPS increased expression levels of inflammatory genes.
The combination of EPA + DHA (1:1) had a more potent anti-inflammatory action than DHA and EPA alone, DHA being a better inhibitor of the studied gene expression than EPA. This implies that giving a mixture of EPA + DHA (1:1) seems to have the best anti-inflammatory effect. The dose of 75 μM was more powerful than 10 and 50 μM. A treatment of T cells with 12.5 μM EPA or DHA during 24 h was done by Verlengia and collaborators . Using the microarray technique, the expression of specific selected genes involved in cytokine production, cell production, signal transduction, and apoptosis was changed. DHA increased the expression of 62% of the studied genes against 33% for EPA. In Raji cells which were treated in the same conditions, 25.9% of the studied genes were regulated by EPA against 8.4% by DHA. Only 3% of the genes were regulated by the two n-3 FAs . These results suggest that molecular mechanisms responsible for the modulatory effect of EPA and DHA on T lymphocytes are different and that DHA regulated a bigger proportion of studied genes. In line with different modulatory effects of EPA and DHA revealed by us and others, an effect of n-3 FAs on heart rate mediated by DHA rather than EPA was previously demonstrated in a supplementation study among humans . High heart rate (HR) has long been associated with CVD morbidity in epidemiological studies. It has been proven that fish oil intake reduced HR mostly in individuals with a high baseline HR and when taken for a long intervention period. In fact, DHA alone (2.8 g/day) diminished HR by 7% in post-menopausal women and DHA but not EPA reduced HR by 3.5 beats per minute and 2.2 beats per minute, respectively, in hyperlipidemic and healthy males .
Regarding cytokine production, it was mostly inhibited with the mixture of EPA + DHA. This inhibition was more pronounced than the one observed with either DHA or EPA alone, DHA being more efficient than EPA. Weldon et al. investigated the action of n-3 FAs on cytokine expression in THP-1 macrophages . Pre-treatment with 100 μM EPA and DHA decreased LPS-stimulated THP-1 cell secretion of TNFA, IL1B, and IL6 compared to control. Even though the effect of the mixture of EPA + DHA was not investigated, the effect of DHA was more important than the one observed with EPA. Similar results were obtained when a lower dose was used, 25 μM DHA decreased the production of IL6 and IL1B more potently than EPA in LPS-stimulated THP-1 macrophages. A similar observation has been reported after addition of EPA and DHA to the culture media of RAW 264.7 macrophages . It must also be noted that in the present study, n-3 FAs decreased IL6 secretion to a greater extent than other cytokines. This corroborates the fact that anti-inflammatory effects of n-3 FAs were cytokine specific rather than global.
To our knowledge, only one study has been designed to provide a head-to-head comparison of the effect of EPA and DHA on inflammation markers as a primary outcome. Allaire and collaborators provided 2.7 g/day of EPA and DHA for 10 weeks to healthy people with abdominal obesity. The group which had DHA supplement had a greater reduction of plasma IL8 levels and greater increase of adiponectin concentration .
The present study showed that in inflammatory conditions, DHA has a better anti-inflammatory effect than EPA. It also showed that the mixture EPA/DHA at a 1:1 ratio more efficiently inhibits inflammation. In human subjects, n-3 FAs have been shown to modulate inflammation-related conditions such as hypertension, dyslipidemia, or insulin resistance . However, it should be noted that, up to now, most trials have not considered that the relative proportion of EPA and DHA may influence the results. This phenomenon might, at least in part, explain the relative controversy on the beneficial effect of n-3 FAs on several conditions. Luis and collaborators  have investigated on Wistar Kyoto rats the effect of a supplementation with different ratios of EPA + DHA on markers of CVD and oxidative stress. EPA + DHA at a 1:1 ratio triggered the most important improvement of the risk factor for type 2 diabetes and reduction of oxidative stress. It is important to point out the fact that in most studies, supplements used have a greater proportion of EPA than DHA. Since DHA appears to have a more important anti-inflammatory action, it might be interesting to reconsider their ratio in supplements, at least for inflammatory diseases.
At the present time, there is no consensus on the ideal n-3 FA intake. Nutritional guidelines have been set by several governments (France, Belgium, Canada) and health organizations (Food and Agriculture Organization, American Dietetic Association) . For instance, the American Heart Association (AHA) recommends that all adults eat fish at least twice a week, which provides 500 mg n-3 FAs/day. It is also recommended for people with documented coronary heart diseases to consume 1 g EPA and DHA/day for secondary prevention . Mozaffarian and collaborators recently reviewed facts concerning EPA and DHA possible shared or complementary effects . In human and animal studies, EPA and DHA reduced platelet aggregation, modulate inflammation, and lower plasma triglyceride (TG) levels. Clinical and observational studies show that DHA increases high-density lipoprotein particles, favors the proportion of large low-density lipoprotein, and causes stronger TG-lowering effects. Mozaffarian et al. concluded that EPA and DHA have complementary and shared benefits. In a study led by Robinson et al., 3T3-L1 adipocytes were incubated in 125 μM of several fatty acids among which are EPA and DHA. Results have shown that DHA increased cellular adiponectin mRNA and secreted adiponectin protein to a greater extent (40% more, P < 0.05) than EPA .
Unfortunately, scientific evidences are still lacking to make quantitative recommendation about the ratio and the dose of EPA and DHA that should be taken to prevent inflammatory diseases. However, it is known that n-3 FAs have anti-inflammatory effects and thus their consumption should be favored. Additional experimental, clinical, and observational investigations are necessary to better understand the complementary and shared effects of EPA and DHA on various clinical outcomes.
A dose effect was obvious for the studied genes; 75 μM of each FA had a stronger anti-inflammatory effect than 10 and 50 μM. It must be noted that a decrease of the expression of all genes except for SOCS1 was seen after the addition of 75 μM n-3 FAs. SOCS1 gene expression was increased. This gene encodes for a suppressor of cytokine signaling. This result suggests that 75 μM but not 10 or 50 μM of EPA, DHA, or EPA + DHA was able to increase SOCS1 expression. SOCS1 being an anti-inflammatory gene, here again, a protective effect of n-3 FAs is showed.
We tested the effect of three different concentrations of each n-3 FA in three different cell culture conditions. In the first condition, the macrophages were incubated in FAs before being stimulated (pre-incubation); in the second, the inflammation and the treatment with n-3 FAs were done at the same time (co-incubation); and in the last condition, inflammation was triggered before adding the FAs (post-incubation). In the post-incubation condition, the anti-inflammatory effect of EPA was greater for IL6, IL1B, TNFA, and TNFAIP3. As far as MCP1 was concerned, the post- and co-incubation had the same effect. EPA + DHA had the best anti-inflammatory effect during the co-incubation for all genes except for SOCS1 where no differences were observed between each condition. Concerning cytokine secretion, the co-incubation seems to be the better condition for EPA, DHA, and EPA + DHA. Globally, findings from gene expression and cytokine production suggest that according to the inflammatory environment, the action of each n-3 FA could be different. Thus, EPA seems to have a greater anti-inflammatory effect on a situation where an inflammatory state is already present; this can be associated to a resolution of inflammation. DHA and the mixture EPA + DHA have a more potent action when they were added at the same time with LPS. This suggests that starting to take EPA and DHA at the same moment with the installation of inflammation might bring out their best anti-inflammatory effects. Even if the anti-inflammatory effect of n-3 FAs on the pre-incubation condition is the less important, a dose effect does exist. It suggests that the protective effect of n-3 FAs among healthy subjects might be visible but less obvious than the one observed among people with pre-existing inflammatory conditions. Current recommendations concerning the consumption of n-3 FAs are available not only as primary prevention of healthy people but also as secondary prevention for people suffering of coronary heart disease or having high TG levels. Human studies investigating the impact of n-3 FA intake against non-CVD were unfortunately not consistent with the data collected in pre-clinical studies. This is why consensus recommendation has not yet been made regarding the possible curative effects of these FAs or their ability to prevent inflammatory disorders. Further investigations are needed to build more precise recommendations which would take into account the body inflammatory state, the best n-3 FAs, and the dose to provide in different inflammatory states.